A diamond-coated face improves a mechanical seal wherever its faces touch: it cuts contact friction several-fold, wears far less than plain silicon carbide, and carries the seal through short losses of lubrication. It does nothing for a seal whose faces already ride on a full fluid film, and no diamond coating makes a seal fit for continuous dry running.
The coating is a diamond film only microns thick, grown by chemical vapour deposition on the running face of a silicon carbide ring.
How Diamond Coated Faces Improve Seal Performance
A diamond coating improves the seal at the contact between its faces: a coated face makes less heat, wears less and survives longer when the fluid film breaks down. Each gain grows with the share of time the faces actually touch.
Lower Friction and Face Heat
In dry contact, a diamond-coated face cuts friction by 83–86% against an uncoated silicon carbide pair.
The chart puts a published dry-friction study, with one ring coated, beside a seal maker’s own test of a seal with both faces coated running in water.
Face heat is the product of friction coefficient, face load and sliding speed, so the heat a seal generates falls in step with the friction. Only vendor data report the resulting drop in face temperature, cooling demand or motor power; no independent test has measured it, so a payback figure rests on the vendor’s own numbers.
Longer Survival When Running Dry
A diamond-coated face survives short dry spells at moderate speed and load, but its dry life is finite and falls steeply as either rises. In the same dry-friction study, the coated face lasted the full three-hour test at low speed and as little as six minutes at the highest load.
Those minutes come from a small ring-on-ring rig, dry, at room temperature, so they rank operating conditions and do not predict field dry-run times. The study gives no dry life for uncoated silicon carbide on the same rig, so no honest “times longer” figure exists.
Resistance to Abrasive Wear
A diamond coating resists abrasive wear far better than plain silicon carbide, but the evidence comes from clean-fluid seal tests; how it fares against solids in the pumped fluid has no independent test behind it. In a 10-day seal wear test, coated faces showed no measurable wear where untreated silicon carbide lost about 0.2 µm.
Which Seal Duties Benefit Most
The duties that gain most are those where the faces spend much of their time in contact: hot water, poorly lubricating fluids, viscous fluids, multiphase fluids, slurries and other abrasive services, and pumps that see dry-running upsets.
Hot water near its boiling point is the clearest case; boiler feed and condensate seals in power generation are typical bad actors of this kind.
Check the failure mode before specifying the coating. It pays on a seal that is dying at the faces, from wear tracks, heat checking or dry-run damage. It will not help a seal failing at the elastomers or springs, or from shaft movement.
Where a Diamond Coating Does Not Help
A diamond coating does not help a seal whose faces already run on a full fluid film, and it does not make a seal fit for continuous dry running; in dry-running flashing hydrocarbons it can add a hazard.
Under a full fluid film the faces do not touch, so a seal that already runs cool and lasts on a well-lubricating product gains nothing from the coating.
An independent dry-running test, commissioned by a seal maker that sells DLC-coated faces, concluded that neither the polycrystalline diamond nor the DLC coating it tested suits extended dry running. A duty that runs dry by design needs face materials built for dry running.
On flashing hydrocarbons the dry-running case needs its own risk assessment: the independent test’s report cites evidence that a polycrystalline diamond face running dry can reach frictional temperatures high enough to bring some of them to auto-ignition.
Diamond adds considerable cost to a face. On a seal that already meets its MTBR target, that premium is wasted; elsewhere the case depends on the gap between current and expected MTBR, with a trial where the duty has no test record.
What to Run Against a Diamond Coated Face
Run a diamond-coated face against plain silicon carbide or against a second diamond-coated face; those are the pairings the published tests favour. Carbon graphite against a coated seat is also sold, but in one published test it reached a lower pressure–velocity limit than diamond against silicon carbide.
Diamond on one or both faces lets silicon carbide run against silicon carbide in the poorly lubricated duties where face material combinations otherwise default to a carbon face.
Coating one ring or both is a trade-off. With one ring coated, the diamond slides against a softer face, which lowers the risk of breakage, delamination and wear; coating both rings is the choice for the most robust seal.
Matching Coating Type and Thickness to the Duty
Coating type matters more than thickness: decide between crystalline diamond and DLC, then insist on a smooth running surface, since none of the published seal tests compared thicknesses by duty.
Crystalline diamond and diamond-like carbon (DLC) are different coatings sold under similar names. Crystalline diamond is grown by CVD on silicon carbide at around 800 °C and is about 5 to 15 µm thick. DLC is largely amorphous, deposited by PVD or plasma-assisted CVD at lower temperatures, and mostly used on steel parts in oil-lubricated systems.
In the independent dry-running test, a plasma-assisted CVD coating of the DLC type ran 18 times longer than the polycrystalline diamond before reaching the test’s temperature limit, and the diamond faces cost three times as much at a similar thickness.
Within crystalline diamond, grain size sets how rough the running surface is, and with it the friction and the wear on the counterface.
| Coating | Grain size | As-grown running surface |
|---|---|---|
| Microcrystalline (MCD) | 1–10 µm | Rough; higher friction and wear |
| Nanocrystalline (NCD) | 100 nm or less | Low roughness |
| Ultrananocrystalline (UNCD) | Under 10 nm | Ra 20–30 nm in the film developed for seal faces |
A multilayer, microcrystalline underneath for adhesion and nanocrystalline on top for a smooth running surface, combines the two. In a published pressurized-water test, multilayer rings sealed while a single microcrystalline layer failed early and never sealed.
The published seal tests used coatings around 10 µm thick, and a supplier’s thickness-by-fluid table is that supplier’s practice.
How Well the Coating Holds Up
A well-bonded coating stays on and keeps sealing, but delamination is a real failure mode, and whether it happens depends on how the coater prepares the base ring.
In the 10-day seal wear test, the ultrananocrystalline film did not delaminate when the silicon carbide face was rougher than 0.1 µm before coating. A ring lapped mirror-smooth is not the best base for a diamond film.
Silicon carbide is the substrate in most of the published work; cobalt-bonded tungsten carbide is harder to make the film adhere to.
A coating can also change the ring’s geometry, and no independent measurement gives a ring’s flatness after diamond coating, so check a coated ring’s flatness in light bands before it goes in.
Plan to replace a worn coated ring rather than repair it: one seal maker states there is no procedure to re-lap or polish its diamond faces.